Mahmoud Srewa

Trustworthy and human-centered AI: federated RLHF, pluralistic LLM alignment, agentic evaluation and coding agents, vision-language-action models for autonomous driving and rehabilitation robotics, and privacy-preserving, robust federated learning.